On Sat, 8 Jan 2011 17:34:34 -0500, Shmuel Metz (Seymour J.) wrote:
>In <[email protected]>, on 01/07/2011
> at 03:36 PM, Binyamin Dissen said:
>
>>This issue is that the binder considers this kind of error to be
>>merely a warning.
>
>What kind of error and why should it be more than a warning?
>
Why should it be less than an Error (RC=8)? Dueling rhetoric. You
say to-mah-to.
Is everyone familiar with this code? I had to look it up:
IEW2556W RECORD NUMBER number OF THE CURRENT OBJECT MODULE HAS AN INVALID
OBJECT IDENTIFIER IN COLUMN 1.
Explanation: Object module records are required to have X'02' in column 1.
System Action: The record in error will be skipped.
In my opinion, there's no plausible etiology that motivates passing this
condition with a Warning. Some languages treat all unrecognizable constructs
as comments. I prefer that Binder not be one of them.
Years ago, on a non-IBM system, I had an object module on punched cards fail
with a similar defect. one card was punched backward. If I inserted it
reversed, it appeared valid, except the wrong corner was cut. I assume
it went through the punch reversed, and a helpful operator spotted the
error and did what he thought was best to correct it.
No more cards, but still no excuse for tolerating the error.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
